From: Eric Niebler (eric_at_[hidden])
Date: 2008-03-22 23:29:24
Larry Evans wrote:
> On 03/22/08 10:29, Larry Evans wrote:
>> On 03/03/08 15:51, Eric Niebler wrote:
>> [snip]
>>> shift_right<L,R>::type is *very* simple. It is expr<tag::shift_right,
>>> args2<L, R> >. Always. It's just a 2-element container and a tag. If L
>>> and R are expressions, it is an expression. If they're grammars, it is a
>>> grammar, suitable for use as the second template parameter to
>>> proto::matches. It is simple, IMO, and leads to a very straightforward
>> So why is there expr<Tag,Args,Arity>::type if it's always the same
>> as its enclosing class, expr<Tag,Args,Arity>. IIRC, mpl requires it
>> for some metaprogramming reason, but it seems that's an implementation
>> detail; consequently, although the user can use it, there's no need
>> for the user to use it. If so, then why do many of the examples have:
>>
>> terminal<...>::type a_term_expr={{}};
>>
>> instead of just:
>>
>> terminal<...> a_term_expr={{}};
>>
>> ?
> OOPS, I misread. I read "shift_right<L,R>::type is shift_right<L,R>"
> not "shift_right<L,R>::type is expr<tag::shift_right,args2<L,R> >".
> I guess I misread because shift_right<L,R> is a grammar; so, I
> assumed that when L and R are grammars, the expr<shift_right,args2<L,R>
> > grammar would be shirt_right<L,R>.
>
> Sorry.
You've got it now. In addition, when L and R are grammars,
shift_right<L,R> and expr<tag::shift_right, args2<L,R> > describe the
same set of expressions, even though they are different types. The first
is the short form for the other.
-- Eric Niebler Boost Consulting www.boost-consulting.com